Natural recolonization of seagrasses may take decades after disturbances and is particularly challenging in near‐shore environments, where sediment mobility inhibits seagrass establishment. We assisted recolonization in fifteen 4x10 m unvegetated experimental plots in a Mexican Caribbean near‐shore fringe where the seagrasses had died due to massive inundations of holopelagic Sargassum species, and where a containment barrier was placed to avoid future inundations. The applied treatments were: artificial substrate (AS: 90 belowground, artificial, biodegradable, 15x15 cm‐sized substrates, cut from a 0.91 x 0.45m Biodegradable EcoSystem Engineering sheet), transplant (TR: 90 cores, 4.5 cm diameter, with Halodule wrightii), and control (C: no manipulation), each with five replicates. After 6 months, 63% of H. wrightii transplants survived, presenting mean rhizome extension of 7.6 cm, and H. wrightii from natural surrounding patches started to colonize the plots. After approximately 8 months, AS and TR plots already showed higher light conditions and lower fluctuations in sediment levels than the controls. After 14 months, the AS and TR plots reached higher mean (± SE) density (respectively, 4024 ± 620 and 3484 ± 360 shoots/m2) and cover (60.3 ± 3.85 and 53.7 ± 2.84 %), compared to the control plots (2043 ± 381 shoots/m2, 35.3 ± 5.7 % cover). Higher density of H. wrightii likely favored the natural establishment of Thalassia testudinum seedlings, with an average of 2.07 (± 0.15) and 1.87 (± 0.18) seedlings in AS and TR plots, respectively, compared to 0.48 (± 0.23) in the controls. Both techniques accelerated seagrass recolonization, but artificial substrates required less effort and avoided harvesting of donor meadows.